INDEX
Negative Logits
oek
-0.08
appoint
-0.08
ao
-0.08
(go
-0.08
empat
-0.07
([[
-0.07
go
-0.07
Nah
-0.07
bless
-0.07
aaaa
-0.07
POSITIVE LOGITS
discard
0.12
reject
0.11
Discard
0.11
discard
0.11
ตรวจ
0.11
invalid
0.10
_invalid
0.10
Invalid
0.10
Quality
0.10
rejecting
0.10
Activations Density 0.005%